https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

1 Recent acceleration of Denman (1972-2017), East , driven by 2 grounding line retreat and changes in configuration.

3

4 Bertie W.J. Miles1*, Jim R. Jordan2, Chris R. Stokes1, Stewart S.R. Jamieson1, G. Hilmar 5 Gudmundsson2, Adrian Jenkins2

6 1Department of Geography, Durham University, Durham, DH1 3LE, UK 7 2Department of Geography and Environmental Sciences, Northumbria University, Newcastle upon Tyne, NE1 8 8ST, UK 9 *Correspondence to: [email protected] 10

11 Abstract: Denman Glacier is one of the largest in , with a catchment that 12 contains an ice volume equivalent to 1.5 m of global sea-level and which sits in the Aurora 13 Subglacial Basin (ASB). Geological evidence of this basin’s sensitivity to past warm periods, 14 combined with recent observations showing that Denman’s ice speed is accelerating, and its 15 grounding line is retreating along a retrograde slope, have raised the prospect that it could 16 contribute to near-future sea-level rise. In this study, we produce the first long-term (~ 50 years) 17 record of past glacier behaviour (ice flow speed, ice tongue structure, and calving) and combine 18 these observations with numerical modelling to explore the likely drivers of its recent change. 19 We find a spatially widespread acceleration of the Denman system since the 1970s across both 20 its grounded (17 ±4% acceleration; 1972-2017) and floating portions (36 ±5% acceleration; 21 1972-2017). Our numerical modelling experiments show that a combination of grounding line 22 retreat, ice tongue thinning and the unpinning of Denman’s ice tongue from a pinning point 23 following its last major calving event are required to simulate an acceleration comparable with 24 observations. Given its bed topography and the geological evidence that Denman Glacier has 25 retreated substantially in the past, its recent grounding line retreat and ice flow acceleration 26 suggest that it could be poised to make a significant contribution to sea level over the coming 27 century.

28

29 1. Introduction

30 Over the past two decades, outlet along the coastline of Wilkes Land, East Antarctica, 31 have been thinning (Pritchard et al., 2009; Flament and Remy, 2012; Helm et al., 2014;

1 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

32 Schröder et al., 2018), losing mass (King et al., 2012; Gardner et al., 2018; Shen et al., 2018; 33 Rignot et al., 2019) and retreating (Miles et al., 2013; Miles et al., 2016). This has raised 34 concerns about the future stability of some major outlet glaciers that primarily drain the Aurora 35 Subglacial Basin (ASB), particularly Totten, Denman, Moscow University and Vanderford 36 Glaciers (Fig. 1). This is because their present day grounding lines are close to deep retrograde 37 slopes (Morlighem et al., 2020), meaning there is clear potential for marine instability 38 and future rapid mass loss (Weertman, 1974; Schoof, 2007), unless ice shelves provide a 39 sufficient buttressing effect (Gudmundsson, 2013). Geological evidence suggests that there 40 may have been substantial retreat of the ice margin in the ASB during the warm interglacials 41 of the Pliocene (Williams et al., 2010; Young et al., 2011; Aitken et al., 2016; Scherer et al., 42 2016), which potentially resulted in global mean sea level contributions of up to 2 m from the 43 ASB (Aitken et al., 2016). This is important because these warm periods of the Pliocene may 44 represent our best analogue for climate by the middle of this century under unmitigated 45 emission trajectories (Burke et al., 2018). Indeed, numerical models now predict future sea 46 level contributions from the outlet glaciers which drain the ASB over the coming decades to 47 centuries (Golledge et al., 2015; Ritz et al., 2015; DeConto and Pollard, 2016), but large 48 uncertainties exist over the magnitude and rates of any future sea level contributions.

49 At present, most studies in Wilkes Land have focused on which is losing mass 50 (Li et al., 2016; Mohajerani et al., 2019) in association with grounding line retreat (Li et al., 51 2015). This has been attributed to wind-forced warm Modified Circumpolar Deep Water 52 accessing the cavity below Totten (Greenbaum et al., 2015; Rintoul et al., 2016; 53 Greene et al., 2017). However, given our most recent understanding of bedrock topography in 54 Wilkes Land, Denman Glacier provides the most direct pathway to the deep interior of the ASB 55 (Gasson et al., 2015; Brancato et al., 2020; Morlighem et al., 2020). Moreover, a recent mass 56 balance estimate (Rignot et al., 2019) has shown that, between 1979 and 2017, Denman 57 Glacier’s catchment may have lost an amount of ice (190 Gt) broadly comparable with Totten 58 Glacier (236 Gt). There have also been several reports of inland thinning of Denman’s fast- 59 flowing trunk (Flament and Remy, 2012; Helm et al., 2014; Young et al., 2015; Schröder et 60 al., 2018) and its grounding line has retreated over the past 20 years (Brancato et al., 2020). 61 However, unlike Totten and other large glaciers which drain marine basins in Antarctica, there 62 has been no detailed study analysing any changes in its calving cycle, velocity or ice tongue 63 structure. This study reports on a range of remote sensing observations from 1962 to 2018 and 64 then brings these observations together with numerical modelling to explore the possible

2 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

65 drivers of Denman's long-term behaviour. The following section outlines the methods (section 66 2) used to generate the remote sensing observations (section 3) and we then outline the 67 numerical modelling experiments (section 4) that were motivated by these observations, 68 followed by the discussion (section 5).

69

70 2. Methods

71 2.1 Ice front and calving cycle reconstruction

72 We use a combination of imagery from the ARGON (1962), Landsat-1 (1972-74), Landsat 4- 73 5 (1989-1991), RADARSAT (1997) and Landsat 7-8 (2000-2018) satellites to create a time 74 series of ice-front position change from 1962-2018. Suitable cloud-free Landsat imagery was 75 first selected using the Google Earth Engine Digitisation Tool (Lea, 2018). Changes in ice- 76 front position were calculated using the box method, which uses an open ended polygon to take 77 into account any uneven changes along the ice-front (Moon and Joughin, 2008). To supplement 78 the large gap in the satellite archive between 1974 and 1989 we use the RESURS KATE-200 79 space-acquired photography from September 1984. This imagery is hosted by the Australian 80 Antarctic Data Centre, and whilst we could not access the full resolution image, the preview 81 image was sufficient to determine the approximate location of the ice-front and confirm that a 82 major calving event took place shortly before the image was acquired (Fig. S1).

83

84 2.2 Velocity

85 Maps of glacier velocity between 1972 and 2002 were created using the COSI-Corr (CO- 86 registration of Optically Sensed Images and Correlation) feature-tracking software (Leprince 87 et al., 2007; Scherler et al., 2008). This requires pairs of cloud-free images where surface 88 features can be identified in both images. We found three suitable image pairs from the older 89 satellite data: Nov 1972 – Feb 1974, Feb 1989 – Nov 1989, and Nov 2001 – Dec 2002. We 90 used a window size of 128 x 128 pixels, before projecting velocities onto a WGS 84 grid at a 91 pixel spacing of 1 km.

92 To reduce noise, we removed all pixels where ice speed was greater than ±50% the MEaSUREs 93 ice velocity product (Rignot et al., 2011b), and all pixels where velocity was <250 m yr-1. Errors 94 are estimated as the sum of the co-registration error (estimated at 1 pixel) and the error in

3 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

95 surface displacement (estimated at 0.5 pixels). This resulted in total errors ranging from 20 to 96 73 m yr-1. Annual estimates of ice speed between 2005-2006 and 2016-2017 were taken from 97 the annual MEaSUREs mosaics (Mouginot et al., 2017). These products are available at a 1 98 km spatial resolution and are created from the stacking of multiple velocity fields from a variety 99 of sensors between July and June in the following year. To produce the ice speed time-series, 100 we extracted the mean value of all pixels within a defined box 10 km behind Denman’s 101 grounding line (see Fig. 3). To eliminate any potential bias from missing pixels, we placed 102 boxes in locations where all pixels were present at each time step.

103 We also estimated changes in the rate of ice-front advance between 1962 and 2018. This is 104 possible because inspection of the imagery reveals that there has been only one major calving 105 event at Denman during this time period because the shape of its ice front remained largely 106 unchanged throughout the observation period. Similar methods have been used elsewhere on 107 ice shelves which have stable ice fronts e.g. Cook East Ice Shelf (Miles et al., 2018). This has 108 the benefit of acting as an independent cross-check on velocities close to the front of the ice 109 tongue that were derived from feature tracking. Whilst the accuracy of velocity fields produced 110 by modern satellites (e.g. Sentinel-1; Landsat-8) have been established, the accuracy of velocity 111 fields produced by earlier, coarser resolution satellite images (e.g. Landsat-1) are largely 112 untested. The ice-front advance rate was calculated by dividing ice-front position change by 113 the number of days between image pairs. Errors are associated with co-registration (1 pixel) 114 and manual mapping of the ice-front (0.5 pixels), resulting in errors ranging from 6 to 73 m yr- 115 1. The general pattern of ice-front advance rates through time is in close agreement with feature 116 tracking-derived changes in velocity over the same time period.

117

118 3. Results

119 3.1 Ice tongue calving cycles and structure

120 Throughout our observational record (1962 - 2018) Denman Glacier underwent only one major 121 calving event, in 1984, which resulted in the formation of a large 54 km long (1,800 km2) 122 tabular iceberg (Fig. 2a-c). Since this calving event in 1984 the ice-front has re-advanced 60 123 km and there have been no further major calving events (Fig. 2a, b), as indicated by minimal 124 changes to the geometry of its 35 km wide ice front. As of November 2018, Denman Glacier’s 125 ice-front was approximately 6 km further advanced than its estimated calving front position 126 immediately prior to the major calving event in 1984 (Fig. 2a, b). However, given the absence

4 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

127 of any significant rifting or structural damage, a calving event in the next few years is unlikely. 128 This suggests the next calving event at Denman will take place from a substantially more 129 advanced position (>10 km) than its last observed event in 1984.

130 Following the production of the large tabular iceberg from Denman Glacier in 1984, it drifted 131 ~60 km northwards before grounding on the sea floor (Fig. 2c), and remained near stationary 132 for 20 years before breaking up and dispersing in 2004. Historical observations of sporadic 133 appearances of a large tabular iceberg in this location in 1840 (Cassin and Wilkes, 1858) and 134 1914 (Mawson, 1915), but not in 1931 (Mawson, 1932), suggest that these low-frequency, 135 high-magnitude calving events are typical of the long-term behaviour of Denman Glacier. In 136 1962, our observations indicate a similar large tabular iceberg was present at the same location 137 (Fig. 2d) and, through extrapolation of the ice-front advance rate between 1962 and 1974 (Fig. 138 2a), we estimate that this iceberg was produced at some point in the mid-1940s. However, the 139 iceberg observed in 1962 (~2,700 km2) was approximately 50% larger in area than the iceberg 140 produced in 1984 (~1,700 km2), and 35% longer (73 km versus 54 km). Thus, whilst Denman’s 141 next calving event will take place from a substantially more advanced position than it did in 142 1984, it may not be unusual in the context of the longer-term behaviour of Denman Glacier 143 (Fig. 2a).

144 There are clear differences in the structure of Denman Glacier between successive calving 145 cycles. In all available satellite imagery between the 1940s and the calving event in 1984 (e.g. 146 1962, 1972 and 1974) an increasing number of rifts (labelled R1 to R7) were observed on its 147 ice tongue throughout this time (Fig. 2e). The rifts periodically form ~10 km inland of 148 Chugunov Island (Fig. 2e), on the western section of the ice tongue, before being advected 149 down-flow. An analysis of the rifting pattern in 1974 and the iceberg formed in 1984 indicates 150 that the iceberg calved from R7 (Fig. 2c, e). In contrast, on both the grounded iceberg observed 151 in 1962 (Fid. 2d), which likely calved in the 1940s, and on the present day calving cycle (1984- 152 present; Fig. 3f), similar rifting patterns are not observed.

153

154 3.2 Ice Speed

155 We observed widespread increases in ice speed across the entire Denman system between 156 1972-74 and 2016-17, with accelerations of 19 ±5% up to 50 km inland of the grounding line 157 along the main trunk of the glacier (Fig. 3a). Specifically, at box D, 10 km inland of the 158 grounding line, ice flow speed increased by 17 ±4% between 1972-74 and 2016-17 (Fig. 3c).

5 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

159 The largest rates of acceleration at box D took place between 1972-74 and 1989 when there 160 was a speed-up of 11 ±5%. Between 1989 and 2016-17 there was a comparatively slower 161 acceleration of 3 ±2% (Fig. 3c). The advance rate of the ice-front followed a similar pattern, 162 but accelerated at a much greater rate. The ice-front advance rate increased by 26 ±5% between 163 1972-74 and 1989, whilst increasing at a slower rate between 1989 and 2018 (9 ±1%; Fig. 3b). 164 At box S on the neighbouring Scott Glacier, we observed a 17 ±10% decrease in velocity 165 between 1972-74 and 2016-17 (Fig. 3d). Similar decreases in ice flow speed are also observed 166 near the shear margin between and Denman Glacier (Fig. 3a, e). The net 167 result of an increase in velocity at Denman Glacier and decreases in velocity either side at the 168 Shackleton Ice Shelf and Scott Glacier is a steepening of the velocity gradient at the shear 169 margins (Fig. 3e). Ice speed profiles across Denman Glacier also indicate lateral migration of 170 the shear margins of ~5 km in both the east and west directions through time (Fig. 3e).

171

172 3.3 Lateral migration of Denman’s ice tongue

173 A comparison of satellite imagery between 1974 and 2002, when Denman’s ice-front was in a 174 similar location (e.g. Fig. 4b, c), reveals a lateral migration of its ice tongue and a change in 175 the characteristics of the shear margins. North of Chugunov Island, towards the ice-front, we 176 observe a bending and westward migration of the ice tongue in 2002, compared to its 1974 177 position (Fig. 4b, c). In 1974, the ice tongue was intensely shearing against Chugunov Island, 178 as indicated by the heavily damaged shear margins (Fig. 4d). However, by 2002 the ice tongue 179 made substantially less contact with Chugunov Island because this section of the ice tongue 180 migrated westwards (Fig. 4d, e). South of Chugunov Island there was a greater divergence of 181 flow between the Denman and Scott Glaciers in 2002 compared to 1974, resulting in a more 182 damaged shear margin (Fig. 4d, e). On the western shear margin between Shackleton Ice Shelf 183 and Denman’s ice tongue there was no obvious change in structure between 1974 and 2002 184 (Fig. 4f, g). However, velocity profiles in this region show an eastward migration of the fast 185 flowing ice tongue (Fig. 3e).

186

187 4. Numerical Modelling

188 4.1. Model Set-Up and Experimental Design

6 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

189 To help assess the possible causes of the acceleration of Denman Glacier since 1972 and the 190 importance of changes we observe on Denman’s ice tongue, we conduct diagnostic numerical 191 modelling experiments using the finite-element, ice dynamics model Úa (Gudmundsson et al, 192 2012). Úa is used to solve the equations of the shallow- or `shelfy-stream’ 193 approximation, (SSA , Cuffey & Paterson, 2010). Previously the model has been used to 194 understand rates and patterns of grounding line migration, and glacier responses to ice shelf 195 buttressing and ice shelf thickness (e.g. Reese et al., 2018; Hill et al., 2019; Gudmundsson et 196 al., 2019), and has been involved in several model intercomparison experiments (e.g. Pattyn et 197 al., 2008; 2012; Leverman et al., 2020).

198 Modelled ice velocities are calculated on a finite-element grid using a vertically-integrated 199 form of the momentum equations. The model domain consists of 93,371 elements with 200 horizontal dimensions ranging from 250 m near the grounding line to 10 km further inland. 201 Zero flow conditions are applied along the inland boundaries, chosen to match zero flow 202 contours from observations. Ice rheology is assumed to follow Glen’s flow law, using stress 203 exponent n=3 and is assumed to follow Weertman’s sliding law, with its own 204 stress exponent, m =3. Other modelling parameters related to ice rheology and basal conditions 205 are the basal slipperiness, C, and the rate factor, A. We initialized the ice-flow model by 206 changing both the ice rate factor A (Fig. S2b) and basal slipperiness C (Fig. S2c), using an 207 inverse approach (Vogel, 2002), iterating until the surface velocities of the numerical model 208 closely matched the 2009 measurements of ice flow (Fig. S2a).

209

210 4.2. Perturbation Experiments

211 We start from a baseline set-up at a fixed point in time where both velocity and ice geometry 212 are well-known. We chose 2009 for this baseline setup, because the calving front is in 213 approximately the same position as in 1972 when our glacier observations start. We use the 214 BedMachine (Morlighem et al., 2020) ice thickness, bathymetry and grounding line position 215 and MEaSUREs ice velocities for 2009 (Mouginot et al., 2017) as inputs. The baseline 216 simulation is then perturbed to test its response to a series of potential drivers that may be 217 responsible for the observed changes in ice geometry since the 1970s. Specifically, we apply 218 observation-based perturbations to test Denman’s response to ice shelf thinning (i), grounding 219 line retreat (ii) and the unpinning of Denman’s ice tongue from Chugunov Island (iii), which 220 are detailed below:

7 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

221 i. To represent ice shelf thinning since 1972, we take the mean annual rate of ice-thickness 222 change from the 1994–2012 ice-shelf thickness change dataset (Paolo et. al., 2015). This annual 223 rate is then applied over the 37 years between 1972 and 2009 to obtain an estimate of the 1972 224 thickness distribution of the Shackleton Ice Shelf, Denman ice tongue and Scott Glacier. We 225 refer to this perturbation as ‘ice shelf thinning’ because the majority of the floating portions of 226 Denman’s ice tongue and Shackleton Ice Shelf have thinned since 1994, although some 227 sections of Scott Glacier have actually thickened near its calving front (Fig. S3).

228 ii. To represent grounding line retreat since 1972 we advanced Denman’s grounding line 229 from its position in the 2009 baseline set-up by 10 km to a possible 1972 position (Fig. S3). 230 We justify a 10 km retreat since 1972 based on the rate of grounding-line retreat observed 231 between 1996 and 2017 (~5km; Brancato et al., 2020). This can be represented in the model 232 by either thickening the ice to ground it, or raising the bed into contact with the bottom of the 233 ice shelf. We have chosen to artificially raise the bedrock to the same height as the bottom of 234 the ice as the less disruptive method, as changing ice thickness would have an effect on ice 235 velocity in addition to that caused by moving the grounding line position. For the newly 236 grounded area, values of the bed slipperiness, C, are not generated in our model inversion, we 237 therefore prescribe nearest-neighbour values to those at the grounding line in the model 238 inversion.

239 iii. To represent the pinning of Denman’s ice tongue against Chugunov Island in the 1972 240 observations (e.g. Fig. 4d, e), we artificially raise a small area of bedrock on the edge of 241 Chugunov Island (Fig. S3). Bed slipperiness was set to a value comparable to that immediately 242 upstream of the grounding line.

243 These three adjustments are applied, both individually and in combination with each other, to 244 the baseline model setup to produce seven different simulations (E1-7) which perturb, 245 respectively:

246 E1. Ice shelf thinning.

247 E2. Grounding line retreat

248 E3. Ice shelf thinning and grounding line retreat

249 E4. Unpinning from Chugunov Island

250 E5. Ice shelf thinning and unpinning from Chugunov Island

8 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

251 E6. Grounding line retreat and unpinning from Chugunov Island

252 E7. Ice shelf thinning, grounding line retreat and the unpinning from Chugunov Island

253 Below we compare the instantaneous change in ice velocity arising from each perturbation 254 experiment to observed changes in velocity, and then use these comparisons to understand the 255 relative importance of each process in contributing to Denman’s behaviour over the past 50 256 years.

257

258 4.3. Model results

259 We show observed 2009 ice speed relative to each simulation which represent possible 1972 260 ice geometries (E1-7, Fig. 5b-h). In all cases, positive (red) values indicate areas where ice was 261 flowing faster and negative (blue) values show areas where ice was flowing slower in 2009 262 relative to each 1972 simulation. Perturbing ice shelf thickness to represent ice shelf thinning 263 since the 1970s results in higher velocities over both the grounded and floating portions of the 264 Denman system (E1, Fig. 5b). However, the simulated acceleration on Denman’s ice tongue 265 (E1, Fig. 5b) is much larger than the observed acceleration, with the simulation showing a 50% 266 acceleration in the area just downstream from the grounding line compared to the observed 267 20% acceleration between 1972 and 2009 (E1, Fig. 5a). Thus, it would appear that ice shelf 268 thinning alone, is not consistent with the observed velocity changes on the Denman system. 269 Perturbing the grounding line to account for a possible grounding line retreat since 1972 270 simulates comparable changes in ice flow speeds to observations near Denman’s grounding 271 line (E2, Fig. 5c), but it is unable to reproduce the observed increases in ice speed across 272 Denman’s ice tongue (E2, Fig. 5c). Thus, grounding line retreat, alone, is also unable to 273 reproduce the observed pattern of velocity changes. Ice shelf thinning and retreating the 274 grounding line results in very similar patterns in ice speed change (E3, Fig. 5d) to that of the 275 grounding line retreat perturbation experiment (E2).

276 In isolation, simulating the unpinning of Denman’s ice tongue from Chugunov Island has a 277 very limited effect on ice flow speeds, with no change in speed near the grounding line and a 278 very spatially limited change on the ice tongue (E4; Fig. 5e). However, when combining the 279 unpinning perturbation with either ice shelf thinning (E5; Fig. 5f) or grounding line retreat (E6; 280 Fig. 5g), it is clear that the unpinning from Chugunov Island causes an acceleration across 281 Denman’s ice tongue. For experiment 5 this results in an even larger overestimate of ice speed

9 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

282 change across Denman’s ice tongue in comparison to experiment 1, which only perturbs ice 283 shelf thickness. However, for experiment 6 the additional inclusion of the unpinning from 284 Chugunov Island to grounding line retreat results in a simulated pattern of ice flow speed 285 change very similar to observations. Specifically the unpinning from Chugunov Island has 286 caused an acceleration across the ice tongue that was not present in experiment 2. Combining 287 all three perturbations (E7, Fig. 5h) produces changes in ice velocity that are most comparable 288 to observations. Both the spatial pattern in ice speed change and the simulated ice speed within 289 box D (Fig. 5i) are very similar to observations for both experiments, and the enhanced 290 westward bending of the directional component of ice velocity in experiment E7 is more 291 consistent with the observed westward bending of the ice tongue (e.g. Fig. 2b).

292

293 5. Discussion

294 5.1 Variation in Denman Glacier’s calving cycle

295 Our calving cycle reconstruction, combined with historical observations (Cassin and Wilkes, 296 1858; Mawson, 1914; 1932) hint that Denman’s multi-decadal high-magnitude calving cycle 297 has remained broadly similar over the past 200 years. It periodically produces a large tabular 298 iceberg, which then drifts ~60 km northwards before grounding on an offshore ridge, and 299 typically remains in place for around 20 years before disintegrating/dispersing. However, more 300 detailed observations and reconstructions of its past three calving events have shown that there 301 are clear differences in both the size of icebergs produced and in ice tongue structure through 302 time (Fig. 2). The large variation (50%) in both the size of iceberg produced and the location 303 the ice front calved from indicates variability in its calving cycle.

304 Extending observational records for ice shelves that calve at irregular intervals, sizes or 305 locations is especially important because it helps to distinguish between changes in glacier 306 dynamics caused by longer-term variations in its calving cycle, and changes in glacier 307 dynamics forced by climate. For example, there have been large variations in ice flow speed at 308 the Brunt Ice Shelf over the past 50 years (Gudmundsson et al., 2017), but these large variations 309 can be explained by internal processes following interactions with local pinning points during 310 the ice shelf’s calving cycle (Gudmundsson et al., 2017). In contrast, the widespread 311 acceleration of outlet glaciers in the Amundsen Sea sector (Mouginot et al., 2014) is linked to 312 enhanced intrusions of warm ocean water increasing basal melt rates (e.g. (Thoma et al., 2008; 313 Jenkins et al., 2018), leading to ice shelf thinning (Paolo et al., 2015) and grounding line retreat

10 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

314 (Rignot et al., 2011a). Thus, in the following section we discuss whether the observed speed- 315 up of Denman since the 1970s (Fig. 3) is more closely linked to variations in its calving cycle 316 (e.g. Brunt Ice Shelf) or if it has been driven by climate and ocean forcing (e.g. Amundsen 317 Sea).

318

319 5.2 What has caused Denman Glacier’s acceleration since the 1970s?

320 We observe a spatially widespread acceleration of both Denman’s floating and grounded ice. 321 This is characterised by a 17 ±4% increase in ice flow speed near the grounding line between 322 1972 and 2017 (Fig. 3c) and a 36 ±5% acceleration in ice-front advance rate from 1972-2017, 323 or 30 ±5% increase in ice-front advance rate between 1962 and 2017 (Fig. 3b). Our estimates 324 of the acceleration in ice front advance rate are of a comparable magnitude to the 36% 325 acceleration of the ice tongue between 1957 and 2017, based on averaged point estimates across 326 the ice tongue from repeat aerial surveys (Dolgushin, 1966; Rignot et al., 2019). Taken 327 together, this suggests a limited change in ice tongue speed between 1957 and 1972, before a 328 rapid acceleration between 1972 and 2017. However, the rate of acceleration throughout this 329 period has not been constant (Fig. 3b, c). From 1972 to 1990, observations indicate that ice 330 accelerated approximately three times faster on the ice tongue (Fig. 3b) and twice as fast at the 331 grounding line (Fig. 3c) in comparison to accelerations in these areas between 1990-2017. 332 When comparing these observations against our numerical modelling experiments we find that 333 a combination of grounding line retreat, changes in ice shelf thickness and the unpinning of ice 334 from Chugunov Island (Fig. 5h) are all required to explain an acceleration of a comparable 335 magnitude and spatial pattern across the Denman system.

336 Averaged basal melt rates across the Shackleton/Denman system are comparable to the Getz 337 Ice Shelf (Depoorter et al., 2013; Rignot et al., 2013). Close to Denman’s deep grounding line, 338 melt rates have been estimated at 45 m yr-1 (Brancato et al., 2020), suggesting the presence of 339 modified Circumpolar Deep Water in the ice shelf cavity. At nearby Totten Glacier (Fig. 1a), 340 wind-driven periodic intrusions of warm water flood the continental shelf and cause increased 341 basal melt rates (Rintoul et al., 2016; Greene et al., 2017) and grounding line retreat (Li et al., 342 2015). It is possible that a similar process may be responsible for some of the observed changes 343 at Denman Glacier. Hydrographic data collected from the Marine Mammals Exploring the 344 Oceans Pole to Pole consortium (Treasure et al., 2017) show water temperatures of -1.31 to - 345 0.26 °C at depths between 550 and 850 m on the continental shelf in front of Denman (Brancato

11 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

346 et al., 2020). Thus, whilst not confirmed, there is clear potential for warm water to reach 347 Denman’s grounding zone and enhance melt rates.

348 Recent observations of grounding line migration at Denman have shown a 5 km retreat along 349 its western flank between 1996 and 2017 (Brancato et al., 2020). However, over this time 350 period there was a limited change in the speed of Denman (2001-2017; 3 ±2% acceleration; 351 Fig. 3c) and our time series indicates that the acceleration initiated earlier, at some point 352 between 1972 and 1990 (Fig. 3c). Reconstructions of the bed topography near the grounding 353 line of Denman Glacier show that the western flank of Denman’s grounding line was resting 354 on a retrograde slope in 1996, a few kilometres behind a topographic ridge (Brancato et al., 355 2020). One possibility is that Denman’s grounding line retreat initiated much earlier at some 356 point in the 1970s in response to increased ocean temperatures enhancing melting of the ice 357 tongue base. This initial grounding line retreat and possible ocean-induced ice tongue thinning 358 may have caused the initial rapid acceleration between 1972 and 1990, before continuing at a 359 slower rate. However, our numerical modelling shows that whilst the combination of the retreat 360 of Denman’s grounding line and ice tongue thinning can produce a similar magnitude of 361 acceleration near the grounding line to observations (E3; Fig. 5e), these modelled processes 362 cannot explain the widespread acceleration across the ice tongue (e.g. E4; Fig. 5d).

363 In order to simulate a comparable spatial acceleration across both Denman’s grounded and 364 floating ice to observations, the un-pinning of ice from Chugunov Island following Denman’s 365 last calving event in 1984 is required (e.g. E6 & 7; Fig. 5g, 5h). In isolation, the reduction in 366 contact with Chugunov Island has had no effect on ice flow speeds at both Denman’s grounding 367 line and ice tongue (E4; Fig. 5d). However, when combined with grounding line retreat and ice 368 tongue thinning, the spatial pattern of simulated ice speed change across the ice tongue more 369 closely resemble observations (E6 & 7; Fig. 5g, 5h). Specifically, the unpinning of the ice 370 tongue from Chugunov Island has caused an acceleration across much of Denman’s ice tongue. 371 The most likely explanation as to why the unpinning from Chugunov Island only influences 372 ice speed patterns in combination with ice tongue thinning and grounding line retreat, and not 373 in isolation, is that ice tongue thinning and grounding line retreat have caused a change in the 374 direction of flow of the ice tongue since the 1970s. In all simulations that perturb either ice 375 tongue thickness or retreat the grounding line (Fig. 5b, c, e, f, g, h), there is a clear westward 376 bending in ice flow direction near Chugunov Island which results in a reduction in contact 377 between the ice tongue and Chugunov Island. This is consistent with observations that show a 378 distinctive westward bending of Denman’s ice tongue since the 1970s (Fig. 2b). These findings

12 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

379 therefore suggest that the reduction in contact with Chugunov Island following Denman’s 380 calving event in 1984 caused an instantaneous acceleration across large sections of its ice 381 tongue, meaning that this calving event has had a direct impact on the spatial pattern of 382 acceleration observed between 1972 and 2017. However, because of the westward bending of 383 Denman’s ice tongue during its re-advance following its 1984 calving event, the ice tongue 384 now makes limited contact with Chugunov Island (e.g. Fig. 4e) and has a very limited effect 385 on ice flow speeds (e.g. E4; Fig. 5e ).

386 The acceleration of Denman’s ice tongue following its last major calving event in 1984 may 387 have also caused a series of positive feedbacks resulting in further acceleration. We observe a 388 steepening of the velocity gradient across Denman’s shear margins and a pattern of the 389 acceleration of the dominant Denman ice tongue and slowdown of the neighbouring Shackleton 390 Ice Shelf and Scott Glacier (Fig. 3a). We also observe the lateral migration of the shear margins 391 at sub-decadal timescales (Fig. 3e). These distinctive patterns in ice speed change are very 392 similar to those reported at the Stamcomb-Wills Ice Shelf (Humbert et al., 2009) and between 393 the Thwaites Ice Tongue and Eastern Ice Shelf (Mouginot et al., 2014; Miles et al., 2020), and 394 are symptomatic of a weakening of shear margins. Therefore, we suggest that at Denman, after 395 the initial acceleration following the reduction in contact with Chugunov Island, the shear 396 margins may have weakened causing further acceleration. We do not include this process in 397 our numerical experiments, and it may explain the divergence between observations and 398 simulated ice speed change in the neighbouring Shackleton Ice Shelf and Scott Glacier (Fig. 399 3a; Fig. 5).

400 Overall, our observations and numerical simulations suggest that the cause of Denman’s 401 acceleration since the 1970s is complex and likely reflects a combination of processes linked 402 to the ocean and a reconfiguration of Denman’s ice tongue. One possibility is that the 403 acceleration of ice across Denman’s grounding line has almost entirely been driven by warm 404 ocean forcing driving grounding line retreat and ice tongue thinning, with the unpinning of 405 Denman’s ice tongue from Chugunov Island only causing a localised acceleration across 406 floating ice. An alternative explanation is that warm ocean forcing has caused ice tongue 407 thinning and grounding line retreat, but the acceleration behind the grounding line has been 408 enhanced through time by changes in ice tongue configuration. Either way, our results highlight 409 that both oceanic processes and the changes in ice tongue structure associated with Denman’s 410 calving event have been important in causing Denman’s observed acceleration.

13 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

411

412 5.3 Future evolution of Denman Glacier

413 In the short-term, an important factor in the evolution of the wider Denman/Shackleton system 414 is Denman’s next calving event. Whilst our observations do not suggest that a calving event is 415 imminent (next 1-2 years), our calving cycle reconstruction indicates that a calving event at 416 some point in the 2020s is highly likely. Because the calving cycle of Denman Glacier has 417 demonstrated some variability in the past (e.g. Fig. 2), the precise geometry of its ice tongue 418 after this calving event cannot be accurately predicted. In particular, it is unclear how 419 Denman’s ice tongue will realign in relation to Chugunov Island following its next calving 420 event. For example, if following Denman’s next calving event the direction of ice flow shifts 421 eastwards to a similar configuration to the 1970s and the ice tongue makes contact with 422 Chugunov Island, the increased resistance could slowdown Denman’s ice tongue for the 423 duration of its calving cycle, but it is unclear if any slowdown could propagate to the grounding 424 line. Thus, this event may have important implications for the evolution of the 425 Denman/Shackleton system for multiple decades.

426 In the medium-term (next 50 years) atmospheric warming could also have a direct impact on 427 the stability of the Denman/Shackleton system. Following the collapse of Larsen B in 2002, 428 Shackleton is now the most northerly major ice shelf remaining in Antarctica, with most of the 429 ice shelf lying outside the Antarctic Circle. Numerous surface meltwater features have been 430 repeatedly reported on its surface (Kingslake et al., 2017; Stokes et al., 2019; Arthur et al., 431 2020). There is no evidence that these features currently have a detrimental impact on its 432 stability, but there is a possibility that projected increases in surface melt (Trusel et al., 2015) 433 could increase the ice shelves vulnerability to meltwater-induced hydrofracturing

434

435 6. Conclusion

436 We have reconstructed Denman Glacier’s calving cycle to show that its previous two calving 437 events (~1940s and 1984) have varied in size by 50% and there have been clear differences in 438 ice tongue structure, with a notable unpinning from Chugunov Island following the 1984 439 calving event. We also observe a long-term acceleration of Denman Glacier across both 440 grounded and floating sections of ice, with both the ice front advance rate and ice near the 441 grounding line accelerating by 36 ±5% and 17 ±4%, respectively, between 1972 and 2017. We

14 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

442 show that in order to simulate a post-1972 acceleration that is comparable with observations, 443 its grounding line must have retreated since the 1970s. We also highlight the importance of the 444 reconfiguration of the Denman ice tongue system in determining the spatial pattern of 445 acceleration observed.

446 The recent changes in the Denman system are important because Denman’s grounding line 447 currently rests on a retrograde slope which extends 50 km into its basin (Morlighem et al., 448 2019; Brancato et al., 2020), suggesting clear potential for marine ice sheet instability. Given 449 the large catchment size, it has potential to make globally significant contributions to mean sea 450 level rise in the coming decades (1.49 m; Morlighem et al., 2020). Crucial to assessing the 451 magnitude of any future sea level contributions is improving our understanding of regional 452 oceanography, and determining whether the observed changes at Denman are the consequence 453 of a longer-term ocean warming. This is in addition to monitoring and understanding the 454 potential impact of any future changes in the complex Shackleton/Denman ice shelf system.

455 In a wider context our results add to the growing body of evidence that some major East 456 Antarctic outlet glaciers, with multi-meter sea-level equivalent catchments have responded to 457 changes in ocean-climate forcing over the past 100 years and, therefore, will be sensitive to 458 projected future warming. The Ninnis Ice Tongue has retreated (Frezzotti et al., 1998), part of 459 the Cook Ice Shelf has collapsed (Miles et al., 2018) and Totten’s grounding line is retreating 460 (Li et al., 2015), and it has been losing mass for several decades (Rignot et al., 2019). However, 461 despite recent improvements (e.g. Morlighem et al., 2020), our understanding of bed 462 topography, oceanography, bathymetry and ice shelf cavity geometry in these key regions of 463 East Antarctic still lag behind that of other regions e.g. Amundsen Sea sector, making the 464 accurate assessment of the magnitude and rates of future sea level contributions challenging.

465

466 Acknowledgements

467 This work was funded by the Natural Environment Research Council (grant number: 468 NE/R000824/1). Landsat and the declassified historical imagery from 1962 is freely available 469 and can be downloaded via Earth Explorer (https://earthexplorer.usgs.gov/). COSI-Corr is 470 available at http://www.tectonics.caltech.edu/slip_history/spot_coseis/download_software 471 .html. The source code for Úa is available at https://doi.org/10.5281/zenodo.3706624. 472 MEaSUREs annual ice velocity maps are available at 473 https://doi.org/10.5067/9T4EPQXTJYW9. We also thank Eric Rignot for providing digitized

15 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

474 estimates of ice flow speed across parts of Denman’s ice tongue, based on the mapped estimates 475 of Dolgushin et al. (1966). 476

477 References

478 Aitken, A. R. A., Roberts, J. L., van Ommen, T. D., Young, D. A., Golledge, N. R., Greenbaum, 479 J. S., Blankenship, D. D., and Siegert, M. J.: Repeated large-scale retreat and advance of 480 Totten Glacier indicated by inland bed erosion, Nature, 533, 385-+, 2016.

481 Arthur, J. F., Stokes, C. R., Jamieson, S. S. R., Carr, J. R., and Leeson, A. A.: Distribution and 482 seasonal evolution of supraglacial lakes on Shackleton Ice Shelf, East Antarctica, The 483 Cryosphere Discuss., https://doi.org/10.5194/tc-2020-101, in review, 2020.

484 Brancato, V., Rignot, E., Milillo, P., Morlighem, M., Mouginot, J., An, L., Scheuchl, B., Jeong, 485 S., Rizzoli, P., Bueso Bello, J. L., and Prats-Iraola, P.: Grounding line retreat of Denman 486 Glacier, East Antarctica, measured with COSMO-SkyMed radar interferometry data, 487 Geophys Res Lett, n/a, e2019GL086291.

488 Burke, K. D., Williams, J. W., Chandler, M. A., Haywood, A. M., Lunt, D. J., and Otto- 489 Bliesner, B. L.: Pliocene and Eocene provide best analogs for near-future climates, P Natl 490 Acad Sci USA, 115, 13288-13293, 2018.

491 Cassin, John, and Charles Wilkes. United States Exploring Expedition: During the Years 1838, 492 1839, 1840, 1841, 1842, Under the Command of Charles Wilkes, USN. Mammalogy and 493 Ornithology. JB Lippincott & Company, 1858.

494 Cuffey, K.M. and W.S.B. Paterson. The physics of glaciers. Fourth edition. Amsterdam, etc., 495 Academic Press. 704pp. ISBN-10: 0-123694-61-2, ISBN-13: 978-0-123-69461-4, 496 hardback, £60.99/€71.95/US$99.95. Journal of , 57(202), 383-384. 497 doi:10.3189/002214311796405906, 2010.

498 DeConto, R. M. and Pollard, D.: Contribution of Antarctica to past and future sea-level rise, 499 Nature, 531, 591-+, 2016.

500 Depoorter, M. A., Bamber, J. L., Griggs, J. A., Lenaerts, J. T. M., Ligtenberg, S. R. M., van 501 den Broeke, M. R., and Moholdt, G.: Calving fluxes and basal melt rates of Antarctic ice 502 shelves, Nature, 502, 89-+, 2013.

503 Dolgushin, L. D. "New data on the rates of movement of Antarctic glaciers." Soviet Antarctic 504 Expedition Information Bulletin 55 (1966): 41-42.

505 Flament, T. and Remy, F.: Dynamic thinning of Antarctic glaciers from along-track repeat 506 radar altimetry, J Glaciol, 58, 830-840, 2012.

507 Frezzotti, M., Cimbelli, A., and Ferrigno, J. G.: Ice-front change and iceberg behaviour along 508 Oates and George V Coasts, Antarctica, 1912-96, Annals of Glaciology, Vol 27, 1998, 27, 509 643-650, 1998.

16 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

510 Gardner, A. S., Moholdt, G., Scambos, T., Fahnstock, M., Ligtenberg, S., van den Broeke, M., 511 and Nilsson, J.: Increased West Antarctic and unchanged East Antarctic ice discharge over 512 the last 7 years, Cryosphere, 12, 521-547, 2018.

513 Gasson, E., DeConto, R., and Pollard, D.: Antarctic bedrock topography uncertainty and ice 514 sheet stability, Geophys Res Lett, 42, 5372-5377, 2015.

515 Golledge, N. R., Kowalewski, D. E., Naish, T. R., Levy, R. H., Fogwill, C. J., and Gasson, E. 516 G. W.: The multi-millennial Antarctic commitment to future sea-level rise, Nature, 526, 517 421-+, 2015.

518 Greenbaum, J. S., Blankenship, D. D., Young, D. A., Richter, T. G., Roberts, J. L., Aitken, A. 519 R. A., Legresy, B., Schroeder, D. M., Warner, R. C., van Ommen, T. D., and Siegert, M. J.: 520 Ocean access to a cavity beneath Totten Glacier in East Antarctica, Nat Geosci, 8, 294-298, 521 2015.

522 Greene, C. A., Blankenship, D. D., Gwyther, D. E., Silvano, A., and van Wijk, E.: Wind causes 523 Totten Ice Shelf melt and acceleration, Science Advances, 3, 2017.

524 Gudmundsson, G. H.: Ice-shelf buttressing and the stability of marine ice sheets, Cryosphere, 525 7, 647-655, 2013.

526 Gudmundsson, G. H., de Rydt, J., and Nagler, T.: Five decades of strong temporal variability 527 in the flow of Brunt Ice Shelf, Antarctica, J Glaciol, 63, 164-175, 2017.

528 Gudmundsson, G. H., Krug, J., Durand, G., Favier, L., and Gagliardini, O.: The stability of 529 grounding lines on retrograde slopes, Cryosphere, 6, 1497-1505, 2012.

530 Gudmundsson, G. H., Paolo, F. S., Adusumilli, S., & Fricker, H. A. Instantaneous Antarctic 531 ice‐ sheet mass loss driven by thinning ice shelves. Geophysical Research Letters, 46, 532 13903– 13909. https://doi.org/10.1029/2019GL085027, 2019

533 Helm, V., Humbert, A., and Miller, H.: Elevation and elevation change of Greenland and 534 Antarctica derived from CryoSat-2, Cryosphere, 8, 1539-1559, 2014.

535 Hill, E. A., Gudmundsson, G. H., Carr, J. R., and Stokes, C. R.: Velocity response of Petermann 536 Glacier, northwest Greenland, to past and future calving events, The Cryosphere, 12, 3907– 537 3921, https://doi.org/10.5194/tc-12-3907-2018, 2018.

538 Humbert, A., Kleiner, T., Mohrholz, C. O., Oelke, C., Greve, R., and Lange, M. A.: A 539 comparative modeling study of the Brunt Ice Shelf/Stancomb-Wills Ice Tongue system, East 540 Antarctica, J Glaciol, 55, 53-65, 2009.

541 Jenkins, A., Shoosmith, D., Dutrieux, P., Jacobs, S., Kim, T. W., Lee, S. H., Ha, H. K., and 542 Stammerjohn, S.: retreat in the Amundsen Sea driven by decadal 543 oceanic variability, Nat Geosci, 11, 733-+, 2018.

17 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

544 King, M. A., Bingham, R. J., Moore, P., Whitehouse, P. L., Bentley, M. J., and Milne, G. A.: 545 Lower satellite-gravimetry estimates of Antarctic sea-level contribution, Nature, 491, 586- 546 +, 2012.

547 Kingslake, J., Ely, J. C., Das, I., and Bell, R. E.: Widespread movement of meltwater onto and 548 across Antarctic ice shelves, Nature, 544, 349-+, 2017.

549 Lea, J. M.: The Google Earth Engine Digitisation Tool (GEEDiT) and the Margin change 550 Quantification Tool (MaQiT) - simple tools for the rapid mapping and quantification of 551 changing Earth surface margins, Earth Surf Dynam, 6, 551-561, 2018.

552 Leprince, S., Ayoub, F., Klinger, Y., and Avouac, J. P.: Co-Registration of Optically Sensed 553 Images and Correlation (COSI-Corr): an operational methodology for ground deformation 554 measurements, Igarss: 2007 Ieee International Geoscience and Remote Sensing 555 Symposium, Vols 1-12, doi: 10.1109/Igarss.2007.4423207, 2007. 1943-+, 2007.

556 Levermann, A., Winkelmann, R., Albrecht, T., Goelzer, H., Golledge, N. R., Greve, R., 557 Huybrechts, P., Jordan, J., Leguy, G., Martin, D., Morlighem, M., Pattyn, F., Pollard, D., 558 Quiquet, A., Rodehacke, C., Seroussi, H., Sutter, J., Zhang, T., Van Breedam, J., Calov, R., 559 DeConto, R., Dumas, C., Garbe, J., Gudmundsson, G. H., Hoffman, M. J., Humbert, A., 560 Kleiner, T., Lipscomb, W. H., Meinshausen, M., Ng, E., Nowicki, S. M. J., Perego, M., 561 Price, S. F., Saito, F., Schlegel, N.-J., Sun, S., and van de Wal, R. S. W.: Projecting 562 Antarctica's contribution to future sea level rise from basal ice shelf melt using linear 563 response functions of 16 ice sheet models (LARMIP-2), Earth Syst. Dynam., 11, 35–76, 564 https://doi.org/10.5194/esd-11-35-2020, 2020.

565 Li, X., Rignot, E., Morlighem, M., Mouginot, J., and Scheuchl, B.: Grounding line retreat of 566 Totten Glacier, East Antarctica, 1996 to 2013, Geophys Res Lett, 42, 8049-8056, 2015.

567 Li, X., Rignot, E., and Mouginot, J.: Ice flow dynamics and mass loss of Totten Glacier, East 568 Antarctica, from 1989 to 2015, Geophys Res Lett, 43, 6366-6373, 2016.

569 Mawson, D., The Home of the Blizzard, Heinemann, London, 1915.

570 Mawson, Douglas. "The BANZ Antarctic Research Expedition, 1929-31." The Geographical 571 Journal 80.2 (1932): 101-126.

572 Miles, B. W. J., Stokes, C. R., and Jamieson, S. S. R.: Pan–ice-sheet change 573 in East Antarctica reveals sensitivity of Wilkes Land to sea-ice changes, Science Advances, 574 2, 2016.

575 Miles, B. W. J., Stokes, C. R., and Jamieson, S. S. R.: Velocity increases at Cook Glacier, East 576 Antarctica, linked to ice shelf loss and a subglacial flood event, Cryosphere, 12, 3123-3136, 577 2018.

18 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

578 Miles, B. W. J., Stokes, C. R., Jenkins, A., Jordan, J. R., Jamieson, S. S. R., and Gudmundsson, 579 G. H.: Intermittent structural weakening and acceleration of the Tongue 580 between 2000 and 2018, J Glaciol, 66, 485-495, 2020.

581 Miles, B. W. J., Stokes, C. R., Vieli, A., and Cox, N. J.: Rapid, climate-driven changes in outlet 582 glaciers on the Pacific coast of East Antarctica, Nature, 500, 563-+, 2013.

583 Mohajerani, Y., Velicogna, I., and Rignot, E.: Evaluation of Regional Climate Models Using 584 Regionally Optimized GRACE Mascons in the Amery and Getz Ice Shelves Basins, 585 Antarctica, Geophys Res Lett, 46, 13883-13891, 2019.

586 Moon, T. and Joughin, I.: Changes in ice front position on Greenland's outlet glaciers from 587 1992 to 2007, J Geophys Res-Earth, 113, 2008.

588 Morlighem, M., Rignot, E., Binder, T., Blankenship, D., Drews, R., Eagles, G., Eisen, O., 589 Ferraccioli, F., Forsberg, R., Fretwell, P., Goel, V., Greenbaum, J. S., Gudmundsson, H., 590 Guo, J. X., Helm, V., Hofstede, C., Howat, I., Humbert, A., Jokat, W., Karlsson, N. B., Lee, 591 W. S., Matsuoka, K., Millan, R., Mouginot, J., Paden, J., Pattyn, F., Roberts, J., Rosier, S., 592 Ruppel, A., Seroussi, H., Smith, E. C., Steinhage, D., Sun, B., van den Broeke, M. R., van 593 Ommen, T. D., van Wessem, M., and Young, D. A.: Deep glacial troughs and stabilizing 594 ridges unveiled beneath the margins of the Antarctic ice sheet, Nat Geosci, 13, 132-+, 2020.

595 Mouginot, J., Rignot, E., and Scheuchl, B.: Sustained increase in ice discharge fromthe 596 Amundsen Sea Embayment, West Antarctica, from1973 to 2013, Geophys Res Lett, 41, 597 1576-1584, 2014.

598 Mouginot, J., Rignot, E., Scheuchl, B., and Millan, R.: Comprehensive Annual Ice Sheet 599 Velocity Mapping Using Landsat-8, Sentinel-1, and RADARSAT-2 Data, Remote Sens- 600 Basel, 9, 2017.

601 Paolo, F. S., Fricker, H. A., and Padman, L.: Volume loss from Antarctic ice shelves is 602 accelerating, Science, doi: 10.1126/science.aaa0940, 2015. 2015.

603 Pritchard, H. D., Arthern, R. J., Vaughan, D. G., and Edwards, L. A.: Extensive dynamic 604 thinning on the margins of the Greenland and Antarctic ice sheets, Nature, 461, 971-975, 605 2009.

606 Pattyn, F., Perichon, L., Aschwanden, A., Breuer, B., de Smedt, B., Gagliardini, O., 607 Gudmundsson, G. H., Hindmarsh, R. C. A., Hubbard, A., Johnson, J. V., Kleiner, T., 608 Konovalov, Y., Martin, C., Payne, A. J., Pollard, D., Price, S., Rückamp, M., Saito, F., 609 Souček, O., Sugiyama, S., and Zwinger, T.: Benchmark experiments for higher-order and 610 full-Stokes ice sheet models (ISMIP–HOM), The Cryosphere, 2, 95–108, 611 https://doi.org/10.5194/tc-2-95-2008, 2008.

612 Pattyn, F., Schoof, C., Perichon, L., Hindmarsh, R. C. A., Bueler, E., de Fleurian, B., Durand, 613 G., Gagliardini, O., Gladstone, R., Goldberg, D., Gudmundsson, G. H., Huybrechts, P., Lee,

19 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

614 V., Nick, F. M., Payne, A. J., Pollard, D., Rybak, O., Saito, F., and Vieli, A.: Results of the 615 Marine Ice Sheet Model Intercomparison Project, MISMIP, The Cryosphere, 6, 573–588, 616 https://doi.org/10.5194/tc-6-573-2012, 2012.

617 Rignot, E., Jacobs, S., Mouginot, J., and Scheuchl, B.: Ice-Shelf Melting Around Antarctica, 618 Science, 341, 266-270, 2013.

619 Rignot, E., Mouginot, J., and Scheuchl, B.: Antarctic grounding line mapping from differential 620 satellite radar interferometry, Geophys Res Lett, 38, 2011a.

621 Rignot, E., Mouginot, J., and Scheuchl, B.: Ice Flow of the Antarctic Ice Sheet, Science, 333, 622 1427-1430, 2011b.

623 Rignot, E., Mouginot, J., Scheuchl, B., van den Broeke, M., van Wessem, M. J., and 624 Morlighem, M.: Four decades of Antarctic Ice Sheet mass balance from 1979-2017, P Natl 625 Acad Sci USA, 116, 1095-1103, 2019.

626 Rintoul, S. R., Silvano, A., Pena-Molino, B., van Wijk, E., Rosenberg, M., Greenbaum, J. S., 627 and Blankenship, D. D.: Ocean heat drives rapid basal melt of the Totten Ice Shelf, Science 628 Advances, 2, 2016.

629 Ritz, C., Edwards, T. L., Durand, G., Payne, A. J., Peyaud, V., and Hindmarsh, R. C. A.: 630 Potential sea-level rise from Antarctic ice-sheet instability constrained by observations, 631 Nature, 528, 115-+, 2015.

632 Scherer, R. P., DeConto, R. M., Pollard, D., and Alley, R. B.: Windblown Pliocene diatoms 633 and East Antarctic Ice Sheet retreat, Nat Commun, 7, 2016.

634 Scherler, D., Leprince, S., and Strecker, M. R.: Glacier-surface velocities in alpine terrain from 635 optical satellite imagery - Accuracy improvement and quality assessment, Remote Sens 636 Environ, 112, 3806-3819, 2008.

637 Schoof, C.: Ice sheet grounding line dynamics: Steady states, stability, and hysteresis, J 638 Geophys Res-Earth, 112, 2007.

639 Schröder, L., Horwath, M., Dietrich, R., and Helm, V.: Four decades of surface elevation 640 change of the Antarctic Ice Sheet from multi-mission satellite altimetry, The Cryosphere 641 Discuss., 2018, 1-25, 2018.

642 Shen, Q., Wang, H. S., Shum, C. K., Jiang, L. M., Hsu, H. T., and Dong, J. L.: Recent high- 643 resolution Antarctic ice velocity maps reveal increased mass loss in Wilkes Land, East 644 Antarctica, Sci Rep-Uk, 8, 2018.

645 Stokes, C. R., Sanderson, J. E., Miles, B. W. J., Jamieson, S. S. R., and Leeson, A. A.: 646 Widespread distribution of supraglacial lakes around the margin of the East Antarctic Ice 647 Sheet. Sci Rep 9, 13823, 2019.

648 Thoma, M., Jenkins, A., Holland, D., and Jacobs, S.: Modelling Circumpolar Deep Water 649 intrusions on the Amundsen Sea continental shelf, Antarctica, Geophys Res Lett, 35, 2008.

20 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

650 Trusel, L. D., Frey, K. E., Das, S. B., Karnauskas, K. B., Munneke, P. K., van Meijgaard, E., 651 and van den Broeke, M. R.: Divergent trajectories of Antarctic surface melt under two 652 twenty-first-century climate scenarios, Nat Geosci, 8, 927-U956, 2015.

653 Vogel, C. R. Computational methods for inverse problems. Vol. 23. Siam, 2002.

654 Williams, T., van de Flierdt, T., Hemming, S. R., Chung, E., Roy, M., and Goldstein, S. L.: 655 Evidence for iceberg armadas from East Antarctica in the Southern Ocean during the late 656 Miocene and early Pliocene, Earth Planet Sc Lett, 290, 351-361, 2010.

657 Young, D. A., Lindzey, L. E., Blankenship, D. D., Greenbaum, J. S., de Gorord, A. G., Kempf, 658 S. D., Roberts, J. L., Warner, R. C., Van Ommen, T., Siegert, M. J., and Le Meur, E.: Land- 659 ice elevation changes from photon-counting swath altimetry: first applications over the 660 Antarctic ice sheet, J Glaciol, 61, 17-28, 2015.

661 Young, D. A., Wright, A. P., Roberts, J. L., Warner, R. C., Young, N. W., Greenbaum, J. S., 662 Schroeder, D. M., Holt, J. W., Sugden, D. E., Blankenship, D. D., van Ommen, T. D., and 663 Siegert, M. J.: A dynamic early East Antarctic Ice Sheet suggested by ice-covered 664 landscapes, Nature, 474, 72-75, 2011.

665 Weertman J., Stability of the junction of an ice sheet and an ice shelf. J. Glaciol., 13(67), 3–11, 666 1974.

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

21 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

683 Figures

684

685 Figure 1: a) Bedrock elevation of the Aurora Subglacial Basin (Morlighem et al., 2020). Note 686 the deep trough inland of the Denman grounding line. b) REMA hill-shade DEM of Denman 687 Glacier, Scott Glacier, and the Shackleton Ice Shelf (Howat et al., 2019). Note the complex 688 composition of the Shackleton Ice Shelf with several pinning points. c) MEaSUREs velocity 689 of Denman Glacier (Rignot et al., 2011) overlain on a Landsat-8 image from November 2018. 690 Note the steep velocity gradient between the Shackleton Ice Shelf, Denman Glacier and Scott 691 Glacier.

22 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

692 693

694

695 Figure 2: a) Reconstructed calving cycle of Denman Glacier 1940-2018. b) Examples of ice- 696 front mapping 1962-2018. Note the change in angle of the ice shelf between its present (light 697 blue – dark blue lines) and previous (pink-red lines) calving cycle. c) and d) Images of the 698 grounded iceberg produced by Denman in 1984 (c) and in the 1940s (d). e) and f) Difference 699 in ice shelf morphology between 1974 and 2015. Note the presence of rifting in e (digitized in 700 black).

701

702

23 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

703

704 Figure 3: a) Percentage difference in ice speed between 2016-17 and 1972-74. Red indicates 705 a relative in increase in 2016-17 and blue a relative decrease in 2016-17. b) Time series of the 706 advance rate of the Denman ice-front 1962-2018. c) Time series of mean ice speed from box 707 D, 1972-2017) approximately 10 km behind the Denman grounding line. d) Time series of 708 mean ice speed from box S, on Scott Glacier, 1972-2017. e) Ice speed profiles across the 709 Shackleton-Denman-Scott system from 1972-74, 1989, 2007-08 and 2016-17.

24 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

710

711 Figure 4: a) Direction of ice tongue migration since 1974. b), d) and f) Close-up in examples 712 of ice tongue structure and location in 1974. c), e) and g) Close-up in examples of ice tongue 713 structure and position in 2002. In particular, note the reduction in contact between Denman 714 Glacier and Chugunov Island between 1974 (d) and 2002 (e).

715

716

717

718

719

720

25 https://doi.org/10.5194/tc-2020-162 Preprint. Discussion started: 6 July 2020 c Author(s) 2020. CC BY 4.0 License.

721 722 Figure 5 – The effect of varying ice geometry on ice flow: Ice velocity difference between 723 2009 observations and a) observations from 1972, and b)-h) seven experiments which perturb 724 2009 ice geometry to represent possible 1972 ice geometry configurations. The seven 725 experiments are: b) ice shelf thinning (E1), c) grounding line retreat (E2), d) ice shelf thinning 726 and grounding line retreat (E3), e) unpinning from Chugunov Island (E4), f) ice shelf thinning 727 and unpinning from Chugunov Island (E5), g) grounding line retreat and unpinning from 728 Chugunov Island (E6) and h) combining ice shelf thinning, grounding line retreat and the 729 unpinning from Chugunov Island (E7). Note that red indicates areas where ice is flowing faster 730 in 2009 and blue indicates areas that are flowing slower with arrows showing the direction and 731 magnitude of change when compared to the 1972 perturbations. Finally, i) the mean speed in 732 box D (located just upstream of the grounding line, shown in black) in the perturbed (1972) 733 configuration, with the observed 1972 and 2009 speed shown in red and blue, respectively. E7 734 most closely matches the speed observed in box D, the spatial pattern of the observed 735 acceleration and the westward bending of Denman’s ice tongue.

26